a baseline study on the effectiveness of a mentor ... · 4.3 cpd certificate programme ... find...
TRANSCRIPT
A Baseline Study on the Effectiveness of a
Mentor Certificate Programme Leading to
Induction Activities for New Teachers in
Rwandan Primary Schools
RESEARCH COMMISSIONED BY VVOB EDUCATION FOR DEVELOPMENT
DO NOT SPREAD OR DISTRIBUTE WITHOUT PERMISSION
Sofie Cabus12 Carla Haelermans3 Ilse Flink4 Andrew Gasozi Ntwali5 Jef Peeraer5
Abstract
Rwanda faces particular challenges with regard to teacher subject and pedagogical content
knowledge and teacher retention. This paper presents a baseline study of an experiment (RCT)
with a mentor certificate programme that should lead to induction activities in Rwanda. These
induction activities imply mentoring and monitoring activities for newly qualified or newly
appointed teachers. Questionnaires were administered among a sample of 719 new teachers
across 6 Districts in Rwanda in March 2019. Of these respondents, 412 teachers have a mentor
that takes part in the mentor certificate programme in the year 2019, developed by VVOB in
partnership with the Rwanda Education Board, and the University of Rwanda College of
Education.
The objective of this baseline study is to assess the baseline status of primary outcome measures
for the intervention and control group and test for any differences between the two groups. The
primary outcome measures are job satisfaction, work related needs satisfaction, job stress and
burnout, teacher efficacy and motivation. Furthermore, the baseline results on these outcome
measures are described.
This baseline study shows that overall treatment and control groups are very similar on pre-
treatment characteristics and on the baseline outcome measures.
Keywords: Baseline Study; Mentor Certificate Programme; Induction Activities; RC.
1 Corresponding author ([email protected]) 2 KU Leuven, Research Institute for Work and Society (HIVA), Parkstraat 47 bus 5300, 3000 Leuven, Belgium. 3 Maastricht University, School of Business and Economics (SBE), PO Box 616, 6200 MD, Maastricht, the
Netherlands. 4 VVOB Rwanda | Education for Development, KG 565 St, Kigali, Rwanda.
2
Table of contents
1 Introduction ................................................................................................................................... 5
2 Literature ....................................................................................................................................... 7
3 Context ........................................................................................................................................... 8
4 Intervention ................................................................................................................................. 11
4.1 Short background ............................................................................................................... 11
4.2 Delivery of the induction programme ............................................................................... 12
4.3 CPD certificate programme ............................................................................................... 13
4.4 Key activities of the induction programme ....................................................................... 15
5 Research method ......................................................................................................................... 16
6 Sampling procedure .................................................................................................................... 17
7 Data .............................................................................................................................................. 19
7.1 Composition of the sample ................................................................................................. 19
7.2 Background characteristics ................................................................................................ 20
7.3 Experience with induction programmes and induction needs ........................................ 22
8 Description and analysis of outcome variables analysis .......................................................... 24
8.1 Properties and reliability .................................................................................................... 24
8.1.1 Job Satisfaction ............................................................................................................. 25
8.1.2 Work Related Needs Satisfaction.................................................................................. 26
8.1.3 Experienced Job Stress and Burnout ............................................................................. 28
8.1.4 Self-Assessed Teacher Efficacy .................................................................................... 29
8.1.5 Motivation toward Work Tasks .................................................................................... 29
8.2 Discussion of quality ........................................................................................................... 30
9 Evaluation of the pre-treatment outcomes ............................................................................... 31
10 Conclusion ............................................................................................................................... 33
11 References ................................................................................................................................ 33
3
List of abbreviations
CBC Competence Based Curriculum
CPD Continuous Professional Development
MT Mentor Trainers
NT New teachers
NT Mentors New teachers’ mentors
NQT Newly qualified teachers: per definition recently graduated from TTC
NAT Newly assigned teachers: not per definition recently graduated from
TTC, but recently assigned to a new school
REB Rwanda Education Board
SBM School Based Mentor
SSL School Subject Leaders
TTC Teacher Training Centre
URCE University of Rwanda - College of Education
4
List of tables
Table 1: T-test on background characteristics ...................................................................................... 21
Table 2: Experience with induction programmes (N=719) ................................................................... 22
Table 3: Participation of respondents in induction activities at their school (N=719) .......................... 23
Table 4: The degree to which respondents currently need formally organized support (N=719)......... 24
Table 5: Scores on the sub-questions of the scale job satisfaction ....................................................... 26
Table 6: Scores on the sub-questions of the scales needs for competence, autonomy and relatedness at
work ...................................................................................................................................................... 27
Table 7: Scores on the sub-questions of the scales personal accomplishments and emotional
exhaustion ............................................................................................................................................. 28
Table 8: Scores on the sub-questions of the scale self-assessed teacher efficacy ................................. 29
Table 9: Scores on the sub-questions of the scales intrinsic motivation; identified regulation;
introjected regulation; external regulation; and amotivation ................................................................ 30
Table 10: T-test of outcome variables ................................................................................................... 32
List of figures
Figure 1: Structure of the Rwandan education system............................................................................ 9
Figure 2: Survival rates in primary education in Rwanda in 2016........................................................ 10
Figure 3: Country map with location of interventions (٭) .................................................................... 13
Figure 4: Design of the RCT ................................................................................................................. 20
Figure 4: Presentation of the components included in the Survey tool ................................................. 25
5
1 Introduction
This paper presents a baseline study of a Randomized Controlled Trial (RCT) with a mentor
certificate programme that should lead to induction activities in Rwanda. Rwanda faces
particular challenges with regard to teacher subject and pedagogical content knowledge. For
example, Maniraho & Christiansen (2015) collected data from 19 teachers from 19 different
Rwandan schools and tested their mathematics content knowledge. The authors indicate poor
results for basic mathematical skills (algebra, geometry and probability/statistics). Further, they
find that teachers’ level of education did not improve math performance.5 To improve the
quality of education and consequently student learning outcomes, The Rwanda Education
Board (REB) introduced a new curriculum in 2015 called the Competence Based Curriculum
(CBC) (REB, 2016). The CBC focuses on knowledge creation and application rather than just
the acquisition of knowledge and skills (Ngendahayo & Askell-Williams, 2016). The CBC has
direct implications for teachers’ teaching practices. It requires teachers to teach differently (i.e.
student-centred teaching methods instead of knowledge-based methods), to work with new
learning materials and to apply new assessment methods. If there is little support for teachers
in adapting these new teaching practices, this is likely to have an impact on teacher motivation
and retention.
One of the main ways that the Rwandan Ministry of Education (MINEDUC) supports
teachers with implementing the CBC is through the appointment of School Based Mentors
(SBM). It is the ultimate goal for every school in Rwanda to have a SBM.6 The SBM plays a
major role in guiding and organising school-based CPD (e.g. coaching and mentoring) and
promoting reflective practice among colleagues.
5 In Rwanda, the majority of teachers obtain a diploma in secondary education in order to enter the teaching
profession. A smaller group of teachers obtain a bachelor diploma. 6 As early as in 2012, SBM were offered and organised by the Rwanda Education Board (REB), however, owing
to recruiting problems, not every school could be offered a SBM.
6
Newly qualified teachers (NQT), or newly appointed teachers (NAT) 7, are of special interest
to the SBM, as they go through a transition period from school-to-work or from school-to-
school. Throughout this paper we will refer to both NQT and NAT as new teachers by using
the joint abbreviation NT. Literature indicates that these transition periods are difficult (among
others, Smith and Ingersoll, 2004). In order to equip NT with the skills necessary to teach the
CBC, a mentor certificate programme was introduced in 2018 as a pilot, in which mentors are
taught to and stimulated to develop and implement induction activities for NT in primary
schools in Rwanda. As of 2019, the programme was rolled out further over more schools. This
programme is developed by VVOB in Rwanda in partnership with REB, the University of
Rwanda College of Education (URCE), and the Ministry of Education (MINEDUC). The main
component of the mentor certificate programme is to simulate induction activities, which
consist of mentoring and monitoring activities at schools (Section 4.4). Furthermore, the
effectiveness of the mentor certificate programme in initiating new induction activities is
monitored by VVOB in Rwanda in collaboration with its partners in Rwanda and the University
of Leuven (Belgium) and Maastricht University (the Netherlands).
The baseline study discussed in this paper refers to the period right before the introduction
of the training and the subsequent new induction activities and indicates the current state of
participants with regards to their job satisfaction, work related needs satisfaction, experienced
job stress, self-assessed efficacy and motivation. In what follows, we discuss previous literature
on induction programmes in Section 2. Section 3 discusses the context of the Rwandan
education system, followed by an overview of the different components of the induction
programme (Section 4). Then, Section 5 presents the research method for the evaluation of the
effectiveness of the induction programme and Section 6 presents the sampling procedure. Data
and descriptive statistics of background characteristics are found in Section 7. A description of
7 Teachers who face a transition to a new school after having had seniority in another school before.
7
the full set of pre-treatment outcome variables is discussed in Section 8. Section 9 evaluates
the outcomes, and Section 10 concludes.
2 Literature
Previous literature indicates that teachers considerably improve performance over the first
three to five years mainly owing to teachers’ informal learning on-the-job (Rockoff, 2004,
2008; Hanushek & Rivkin, 2010; Hanushek, 2011). However, in these first critical years in-
service, New Teachers (NT) are also more likely to leave the profession early. There are several
factors that contribute to teacher attrition: (1) teachers can face a difficult integration into the
new school; (2) there may be system constraints, or difficulties in finding a permanent job at
the new school (i.e. the problem of frequent migration of teachers), or salary and status issues,
as in some countries, like Rwanda, teachers have very low status and/or salaries.
These reasons, in turn, may negatively affect teaching practices and teacher efficacy of NT,
which in turn may reduce teachers’ value added, motivation and overall job satisfaction, and
increase stress, emotional exhaustion and burnout (among others; Hackman & Oldham, 1975;
Broeck et al., 2010; Strong, 2005; Richter et al., 2013; Cabus et al., 2019). These negative
feelings towards the teaching profession can, in the worst case, trigger emotional exhaustion
and burnout (Richter et al., 2013).
Smith & Ingersoll (2004) and Ingersoll & Strong (2011) discuss the important role for
induction programmes to stimulate teachers’ competences and to retain them on-the-job within
the first years of employment. Induction programmes consist of activities for beginning
teachers, mostly organised by the school and/or in cooperation with the pre-service teacher
training. Examples are: mentoring and coaching, video-analysis of teachers’ teaching, and peer
observation and discussions (Maandag et al., 2007; Algozzine et al., 2007; Davis & Higdon,
2008; Roehrig et al., 2008; Wang et al., 2008).
8
From the literature it is clear that especially mentoring activities are important for teacher
retention. A mentor is often a more experienced colleague at the school of the new teacher
(Ingersoll and Smith, 2004).8 This mentor offers support to the new teacher in, for example,
subject or pedagogical content knowledge and class management, and introduces the new
teacher to the school and his/her new colleagues. However, there are some conditions attached
to who the mentor is, for example: a mentor should be a person that the new teacher can trust;
the chosen mentor should stay appointed for more than one year; and the mentor should have
a professional attitude towards the mentoring activities (Evertson & Smithey, 2000; Cohen &
Fuller, 2006; Fletcher & Strong, 2009; Smith & Ingersoll, 2004; Strong, 2005; Odell & Huling,
2000; Stanulis & Floden, 2009). Further, there are requirements attached to the assignment of
the mentor. Based on a systematic literature study, Cabus et al. (2019) argue that mentoring
programmes are more effective in case of higher frequency, longer duration, and better quality.
But how frequent and how long a mentoring programme should run in order to be effective
remains largely unclear. With regard to mentoring quality, Richter et al. (2013) indicate that
mentoring and coaching of NT should aim at knowledge transformation rather than at
knowledge transmission. Knowledge transformation is defined as a learning process that
attaches new knowledge to prior knowledge of the mentee, in this case the new teacher.
3 Context
Similar to other countries, the formal education system in Rwanda has four main levels or
sub-sectors: pre-primary, primary, secondary, and higher education. This report focuses
particularly on primary education.
8 According to the definition of Ingersoll and Smith (2004, p.683), mentoring is defined as: “[…] the personal
guidance provided, usually by seasoned veterans, to beginning teachers in schools.” We elaborate on this
definition by also including experienced teachers into the target group for mentoring activities, who recently
started in a new school.
9
Compulsory education takes 12 years, where children's age normally ranges from 7 to 18.
Compulsory education covers primary education, lower and upper secondary education
(MINEDUC, 2017). There are national examinations at the end of primary, lower secondary
and upper secondary education that determine eligibility for proceeding to the next level of
education. The language of instruction in primary grade 1 to 3 (P1-P3) is Kinyarwanda, and
changes to English as of primary grade 4 (P4). Figure 1 shows the structure of the Rwandan
formal education system, including pre-primary education.
Figure 1: Structure of the Rwandan education system
Source: Figure 1 – Structure of the Rwandan educations system in MINEDUC (2017)
Education Sector Strategic Plan 2018/19 to 2022/23 final draft 3rd October 2017, pp 14.
Because Rwanda is committed to providing free education for all children aged 7-18, access
to education has increased significantly in the early 2000s. Accordingly, enrolment rates in
primary education in Rwanda have been quite high since 2012, with 97.7% of children enrolled
in P1 in 2016 (MINEDUC, 2017). However, many children in primary education are not within
10
the expected age range owing to grade repetition and dropout. Figure 2 plots the survival rates
(Y-axis) over the primary school grades (X-axis) and indicates that by primary grade 6 (P6)
only about 50 percent of children are still in school. This implies that throughout their primary
school career, many primary pupils drop-out.
Figure 2: Survival rates in primary education in Rwanda in 2016.
Source: Figure 5 in MINEDUC (2017) Education Sector Strategic Plan 2018/19 to 2022/23
final draft 3rd October 2017, pp 22.
Although access rates to primary education are quite high, the quality of education remains
a serious challenge. We identify three indicators of the quality of education: learners’ basic
literacy and numeracy skills, pupil-teacher ratio, and the proportion of adequately trained
teachers. First, quality can be expressed in mastering basic literacy and numeracy skills. In fact,
to successfully progress through primary, secondary and tertiary education, and to be able to
master technical and vocational courses, it is necessary to have a strong basis in literacy and
numeracy, which is not the case for many students. In 2011 and 2014 an assessment of the
learning achievements was conducted in Rwandan schools in order to provide an overview of
the levels of learning in literacy and numeracy of children in primary grade 2 (P2) and primary
grade 5 (P5). In 2014, the results showed that respectively 45.3% and 32.9% of P2 students
achieved the grade level competency in literacy and numeracy. In P5, these proportions largely
11
remained unchanged ( 44.1% and 38.3% in P5, for literacy and numeracy, respectively). Pupils
from urban schools did perform better than those in rural schools, and learners in private
schools performed better than their peers in public schools (government schools) (MINEDUC,
2017).
Second, the pupil-teacher ratio in primary education in Rwanda – an often used indicator of
the quality of education (Glewwe et al., 2011) - is as high as 58:1. What is more, the ratio of
pupils per qualified teacher is 62:1 (with a target of 48:1 for 2017/18) (MINEDUC, 2017). Note
nonetheless, that there is large variation in the average class size and pupil-teacher ratios across
and within districts within the country. To lower the pupil-teacher ratio, Rwanda has introduced
the double-shift system. The double-shift practice entails that teachers educate a different group
of students in the mornings and afternoons, implying that students only get education half a
day, but teachers work full day shifts.
Third, the quality of education can be indicated by the number of adequately qualified
teachers in service (MINEDUC, 2017). One of the key educational challenges in Rwanda is
that teachers lack the required competencies in subject content, pedagogy and language of
instruction (English and Kinyarwanda). About 93% of teachers is qualified, however, in
practice quality between teachers varies (MINEDUC, 2017). We also present evidence on this
in Section 7.
4 Intervention
4.1 Short background
In close collaboration with REB, the Ministry of Education, and URCE, VVOB introduced
a pilot of a mentor certificate programme in 2018, which subsequently would have to lead to
new induction activities for NT (NQT and NAT) in primary education in Rwanda. This pilot
was simultaneously researched, and in 2019 further rolled out in more schools, and again
researched. NQTs are in the first 3 years of their teaching career after leaving pre-service
12
teacher education, while NATs have a maximum of 3 years teaching experience in their newly
assigned school. The induction activities that the mentor certificate programme focused on
consist of mentoring and monitoring activities. The mentoring activities consist of an individual
learning plan that is put together by the NT and his/her mentor. The learning plan is essentially
a bundle of activities that include formal one-to-one mentoring (e.g. on pedagogical and subject
content knowledge), and discussions with peers through a community of practice. These
activities take place on average two times per week. Then, three times per year, a tutor of the
pre-service teacher training centre (TTC) monitors the NT by observing his/her teaching
practice and providing feedback. Potential problems with the mentor can also be discussed with
the TTC tutor.
4.2 Delivery of the induction programme
This implementation of the mentor certificate programme and subsequent induction
programme started with a pilot in the year 2018. To measure the effectiveness of the pilot
programme, VVOB, URCE, REB and the Universities of Leuven and Maastricht established a
research programme. In 2019, the programme was rolled out in a larger amount of schools. The
induction activities at school are organized by School-Based Mentors (SBM), School-Subject
Leaders (SSL) and Tutors from the Teacher Training College (TTC), who are all trained in
educational coaching and mentoring through certified CPD Programmes from the University
of Rwanda College of Education, co-developed and -implemented with VVOB. More
information on the CPD certificate programmes is provided in section 4.3.
The induction programme is implemented in 6 districts in the Western Province (Nyabihu
and Rusizi) and Eastern Province (Nyagatare, Gatsibo, Kirehe and Kayonza) of Rwanda
Figure 3). These districts have higher dropout rates and lower examination results than on
average observed in Rwanda (Ministry of Education, 2016) (as can be seen in Figure 3 based
13
on the colours; red districts have high drop-out rates). In total 275 school-based mentors are
trained in the year 2019, implying that the induction programme will be introduced to all NTs
in these primary (intervention) schools in Rwanda.
Figure 3: Country map with location of interventions (٭)
Source: 2015 Educational Statistical Yearbook, Ministry of Education (2016), p.31
4.3 CPD certificate programme
The support and guidance (including mentoring) provided to NTs in the induction
programme is about building the teaching profession, retaining teachers in the profession, and
ensuring that they are part of a learning community focused on continuously improving their
teaching practice and their learning. The induction programme should be considered part of the
School-Based Mentorship Programme Framework (SBMPF). A School-Based Mentor (SBM)
٭
٭
٭
٭
٭
٭
14
has been selected in each public/government aided school to facilitate the SBMPF
implementation.
In this context, VVOB, REB, and URCE, have developed a CPD certificate programme for
School-Based Mentors (SBM), School-Subject Leaders (SSL) in mathematics and TTC tutors.
TTC tutors co-facilitate the delivery of induction programme and support SBMs and SSLs
(NTs mentors in schools) in their mentoring of NTs. The TTC tutor also establishes a link
between the pre-service and in-service teaching training. This is in line with Policy Priority 4
of the draft TDM-policy, which stipulates that CPD of teachers and pre-service teacher
education must be more closely linked.
The CPD certificate programme on Educational Mentorship and Coaching provides SBMs
with coaching and mentoring skills needed to guide and organize CPD at their school, to
promote reflective practice in their respective schools and to advance the implementation of
the CBC. It focuses on teacher development as an ongoing process in a teacher’s career.
Throughout this paper, the person, who has followed the Educational Mentorship and Coaching
programme, will be called the new teachers’ (NT) mentor.
There is also a CPD certificate programme for SSL in mathematics. The programme aims to
boost basic mathematical skills among teachers, and is essentially a certificate course for SSL
in mathematics. SSL are introduced to a variety of aspects of pedagogical content knowledge
on mathematics and mathematical leadership. It also contains examples related to selected
topics of the primary mathematics curriculum, in order to mentor NT who teach mathematics.
TTC tutors also follow the CPD certificate programme on Educational Mentorship and
Coaching organised by REB. They receive additional training on monitoring the performance
of NT. In addition, TTC tutors will be coached by the URCE in their role as co-trainer and
coach of mentors (SSL and SBM).
15
Given that SBMs and TTC tutors are the main facilitators of induction programmes in
schools, the RCT focuses on mentoring by SBM and TTC tutors, not by the SSL.
4.4 Key activities of the induction programme
Based on research evidence on means for effective CPD, elements and success factors of
induction and in alignment with the priorities and strategies of the Government of Rwanda, the
induction programme in Rwandan schools should consist of four key activities, which are
covered in the CPD certificate programmes. The first main activity is mentoring, which can be
both one-to-one mentoring and group mentoring. Examples of mentoring activities are joint
lesson planning, observing mentor’s teaching or observing fellow teachers - including a
discussion before (preparation lesson observation) and after observation (debriefing and
reflection) - analysing students’ work and results on assessments, analysing marking and record
keeping system, discussing about teaching and learning issues, or suggesting and discussing
teaching and classroom management techniques. The second activity is CoP sessions, where
NT and more experienced teachers, teaching the same subject (depending on the size of the
school) meet to discuss their work. They think of solutions to challenges they encounter in the
classroom and share good practices. Activities that can be undertaken during a CoP-session
are, for example, collaborative lesson preparation, lesson study/observation, case discussions,
analysing student work on formative and summative assessments, analysing marking and
record keeping systems, or developing strategies for teaching learners with special educational
needs (SEN). The third activity are seminars or trainings on topics of concern to new teachers,
which can be organised at school, sector or district level. The fourth and final activity is TTC
support to NTs’ induction. This can include classroom observation, analysing students’ work
and results on assessments or reviewing a NT’s CPD-plan (review progress and setting targets)
and monitoring the implementation of the induction programmes.
16
Note that the intervention technically is an intent-to-treat for the NTs, as, the SBMs and SSLs
decide which activities actually to implement in their schools, meaning that the content,
frequency and intensity of the actual activities within the induction programme vary per school,
and per NT.
5 Research method
During the pilot study in 2018, we have set-up and tested a valuable survey tool for the
measurement of the effectiveness of the intervention “induction programmes for New Teachers
in Rwanda”. The survey tool measures different teacher outcomes that are expected to change
when NT have access to well-designed induction programmes, namely: job satisfaction,
motivation, teaching practices, teacher efficacy, and stress and burnout. The measurement
instruments underlying these outcomes are discussed in Section 8. In order to measure the
effectiveness of the intervention programme in terms of the aforementioned teacher outcomes,
we wish to compare a treatment group with a control group before and after the intervention
has taken place. To this end, we will eventually conduct a difference-in-differences (DiD)
analysis. However, in this baseline report we can obviously only report on the pre-treatment
measures, and the differences of treatment and control groups on these measures.
However, two different conditions apply to the treatment group. A first treatment group
receives mentoring activities within the scope of the new induction programme. In the second
treatment group a TTC tutor monitors the NTs’ performance in addition to the mentoring
activities. Therefore, in the remainder of this paper, we distinguish between two treatment
groups, namely: the mentoring group (Treatment 1) and the mentoring and monitoring group
(Treatment 2).
There are a couple of assumptions underlying a DiD analysis. First, it is important that the
intervention group and the control group are comparable based on observed and unobserved
background characteristics. For example, teachers, who participate in the intervention should
17
have a similar intrinsic motivation for teaching as teachers in the control group. If this condition
is met, the effectiveness of the intervention can be ascribed to the intervention and not to
differences in characteristics between the treated and untreated groups.
Second, the intervention group and the control group move “parallel over time”. This parallel
time trend assumption implies, for example, that teachers from the intervention have a similar
likelihood to drop-out from the teaching profession than teachers from the control group. In
other words: before the intervention has taken place, treated teachers have similar outcomes
than untreated teachers.
And third, there is no overlap (or ‘spill-over’) between participants and non-participants to
the intervention. Spill-over effects violate the stable unit treatment value assumption
(SUTVA); which implies that untreated teachers benefit from the intervention through access
to knowledge of or experiences with induction programmes via treated teachers.
A good way for ensuring comparability between the treated and untreated teachers, and to
meet the parallel time trend assumption, is random assignment (to the intervention). Random
assignment means that differences between treated and untreated teachers are based on a
random error in the assignment process. The idea is that these random errors in observed or
unobserved characteristics cancel each other out in the comparison of the intervention and
control group before and after the intervention, so that the estimated effectiveness of induction
programmes is a ‘true’ measure of the effect.
6 Sampling procedure
Prior to the pilot study in 2018, all 589 schools/SBMs that were eligible to participate in the
CPD certificate programme on educational mentorship and coaching were randomly assigned
to three training cohorts. The first training cohort was considered the pilot group which was
18
trained in 2018 and included N=38 schools/SBMs. The second cohort, considered the treatment
group which is trained in 2019, included N=275 schools/SBMs. The last cohort, considered the
control group in this study in 2019 which will be trained in 2020, included N=276
schools/SBMs. As such, a total of 551 schools were deemed eligible to participate in the RCT.
As funds did not allow to select all 551 schools for the RCT, we drew random samples from
each of the two groups/cohorts. To select a sample large enough to detect a small effect of the
treatment we used G-power software to calculate the sample size according to the number of
NT. Based on these results, the sample size was set at 1,050 NTs, enabling the estimation of a
significant small standardized effect of 0.2 (Cohen’s d) with a statistical power of 0.80
(convention). Calculating this back to the number of schools, this implied selecting 231 schools
(assuming at least 4 NTs per school, the average number of NTs at each pilot school). As having
two treatment groups required us to have 3/2 allocation ratio for the treatment versus the control
group, we selected 124 treatment schools and 106 control schools for the RCT. The treatment
sample is composed of a random selection of 72 schools that are part of treatment 1 and all 52
schools that are part of treatment 2. The 106 control schools were randomly selected from the
sample of 276 schools.
Next, enumerators went to the selected schools to interview all the NT. Given that some
schools had less NT than expected, no NT at all and/or some NT did not want to participate in
the survey, the final sample, and after data cleaning, included a total of 214 schools and 719
NT, which is a negligible difference from the 231 schools that we needed according to the
power analysis. This includes 275 NT in treatment group 1, 137 NT in treatment group 2 and
307 NT in the control group. Figure 4 gives an overview of the design of the RCT, leading
from the target population of schools to the sample that is used for analysis. In order to meet
the third assumption on “no spill over effects” between treated and untreated teachers, we
guaranteed that treated teachers are not teaching in control schools, and vice versa.
19
7 Data
7.1 Composition of the sample
While 760 teachers were surveyed in March 2019, the current dataset consists of 719 teachers
after cleaning, since we lose 41 observations due to missing information on personal
background characteristics of teachers. In total the population for analysis included, 212
schools: 71 schools (or 275 NT) in Treatment group 1; 48 schools (137 NT) in Treatment Group
2; and 93 schools (307 NT) in the Control Group. The total sample size is then equal to N=719
new teachers.
20
Figure 4: Design of the RCT
Note: This figure represents the total number of schools and the total number of NT in the final dataset, after
data cleaning. Before data cleaning, we had questionnaires among 760 NT or 116 schools.
7.2 Background characteristics
Table 1 summarizes the background characteristics of the sample of teachers, and the
differences between the two treatment groups and the control group. We have slightly more
men than women in the sample. They are on average 28 years old. Age is not statistically
different between the two treatment groups and the untreated teachers. However, gender
appears to be significantly different between the second treatment group and the control group.
There are 10 percentage points more males in the control group than in Treatment Group 1.
Educational attainment is rather poor. In the control group, we observe that 91.5 percent of
teachers attained level-1. This percentage share is equal to 85.1 percent in Treatment Group 1
(which is significantly different at the 5% -level), and 87.6 percent in Treatment Group 2 (not
statistically different). However, since we apply eight consecutive t-tests, we have to be
cautious with the interpretation, as by statistical law we would expect every 1 out of 20 tests to
be significant if we use a 5% significance level9, and indeed, we see some significant
differences a the 10%-level.
9 In some studies authors apply a Bonferroni correction with a large number of t-tests. However, the Bonferroni
correction is a rather strict test that is undesirable and is unusual to apply to pre-treatment comparisons, as it
means to account for false positives in outcomes.
21
All teachers teach in primary school, but 130 teachers (18%) also teach in lower secondary
education. The share of teachers also teaching in secondary education varies highly by district
(varying from 3 to 37 percent, with an average of 18), but within districts treatment and control
groups are comparable. On average, respondents travel thirty to forty minutes to the appointed
school. We observe small significant differences between the treatment groups compared to
the control group with regard to travel time. However, the difference between the groups is
very small (5 to 7 minutes).
Table 1: T-test on background characteristics
Control group Treatment Group 1 Sig. Treatment Group 2 Sig.
Mean Std.Dev Mean Std.Dev Mean Std.Dev
male (%) 0.570 0.496 0.564 0.497 0.474 0.501 *
age (years) 27.7 5.6 28.0 6.0 27.9 6.5 teacher in primary
vs. secondary (%) 0.84 0.37 0.78 0.41 * 0.85 0.36
educational attainment (%)
level-1 0.915 0.279 0.851 0.357 ** 0.876 0.331
level-2 0.049 0.216 0.065 0.248 0.058 0.235
level-3 0.036 0.186 0.076 0.266 ** 0.066 0.249
level-4 0.000 0.000 0.007 0.085 0.000 0.000
teaching experience (years)
since graduation 3.4 4.0 3.5 0.3 3.4 3.9
in this school 1.5 1.0 1.3 0.1 ** 1.4 1.0
Travel time (minutes) 39.6 36.7 34.0 30.3 ** 32.5 26.8 **
Note: Stars denote the significant levels of the differences between the treatment group (1/2) and the control
group. Significant levels are denoted at 1%-level (***); 5%-level (**); and 10%-level (*). Treatment Group 1
receives mentoring only (N=275); and Treatment Group 2 receives mentoring and monitoring (N=137).
Furthermore, about 415 teachers (57.7%) have had teaching experience in no more than one
school. Years of teaching experience since graduation from pre-service teacher training varies
from 0 to 31 years with an average of three and a half years and a standard deviation of 4. As
such, there are a few teachers with a lot of seniority, however, they got appointed to a new
school fairly recently. There is no significant difference in this respect between the control
group and both the treatment groups.
22
7.3 Experience with induction programmes and induction needs
One of the targets of the Mentor Certificate Programme is that it could lead to induction
activities for New Teachers in Rwandan Primary Schools. In our questionnaire, we have
explored more explicitly whether teachers have experiences with induction programmes, or
express certain induction needs (Table 2, Table 3 and Table 4). Of the full sample of teachers,
38 percent responds to be familiar with the definition of induction programmes; 54 percent has
a new teacher mentor at the school; and 17 percent has a TTC mentor. The new teacher mentor
is mostly a school-based mentor (41%), but can also be the director of studies or deputy head
teacher (26%); or the school subject leader (24%).10 Those respondents with a new teacher
mentor talk very frequently with him/her, for example, every month (23%); every two weeks
(10%); once a week (24%); several times per week (15%) or on a daily basis (6%). Only 10
percent of respondents talk less than once a month with the new teacher mentor, and 12 percent
never.
Table 2: Experience with induction programmes (N=719)
Question Yes No
Are you familiar with the definition of induction programmes for new teachers? 38% 62%
I took part in a professional mentoring programme at my school 29% 71%
There was an organized introduction in my school 27% 73%
I took part in general and/or administrative introduction to my school 83% 17%
Do you have a new teacher mentor at your school? 54% 46%
Who is the new teacher mentor?* School subject leader 24% Director of studies/deputy head teacher 26% school based mentor 41% Other 10% Do you have a TTC mentor? 17% 83%
*Note: Answers to this question only provided for those respondents that answered ‘yes, I have a new teacher
mentor at my school.’
10 The remaining group of respondents (9%) answered that the new teacher mentor is “other”
than previously defined persons.
23
More than 80 percent of teachers received a general and/or administrative introduction.
However, barely one in every three respondents took part in a professional mentoring
programme or an organized introduction at the school. This is also indicated by the teachers’
answers on participation in activities at school. Over two-thirds of respondents have never had
any activity that resorts to an induction programme, like joint lesson planning together with a
mentor, classroom observation, individual coaching, developing and reviewing a continuous
professional development plan (CPD), peer discussions or workshops on (the didactics of) their
subject.
Table 3: Participation of respondents in induction activities at their school (N=719)
Never
Less
than
once a
month
Once a
month
Every
two
weeks
Once a
week
Several
times a
week
Every
day
Joint lesson planning together with my
mentor 59% 7% 12% 5% 8% 6% 3%
Observing mentor teaching 67% 6% 11% 3% 7% 4% 1%
Observing another teacher teaching 39% 13% 18% 6% 11% 8% 4%
Being observed by my mentor 48% 9% 21% 6% 9% 7% 0%
Being observed by another teacher 45% 10% 18% 5% 9% 9% 4%
Individual coaching conversation with
my mentor 45% 9% 19% 6% 11% 7% 3%
Developing my individual CPD Plan
together with my mentor 58% 7% 13% 6% 10% 5% 2%
Reviewing my individual CPD Plan
together with my mentor 56% 8% 13% 5% 10% 5% 3%
Being part of a Community of Practice of
teachers teaching the same subject 49% 12% 15% 5% 11% 6% 2%
Being part of a Community of Practice of
teachers teaching different subject 46% 15% 18% 7% 8% 5% 1%
Seminars/trainings on topics of concern
to new teachers 74% 8% 8% 3% 4% 2% 0%
Analysing student work and results on
assessments together with my mentor 60% 8% 14% 4% 7% 5% 2%
Despite this low participation in professional mentoring activities, there is clearly a need for
it (Table 4). For example, respondents wish to have formally organised support at the school
in order to learn about strategies to manage large classes. Further, teachers indicate that they
could benefit from organised workshops, peer discussions or discussions with a mentor at
24
school on didactic material, knowledge and understanding of (teaching) their subject, and
administrative documents. Need for formally organised support seems to be lowest for support
with student evaluation and assessment and student behaviour.
Table 4: The degree to which respondents currently need formally organized support
(N=719)
No need at
present
Low level of
need
Moderate level of
need
High level of
need
Knowledge and understanding of my
subject(s) 13% 22% 37% 28%
Pedagogical competences in teaching my
subject(s) 12% 28% 35% 26%
Student evaluation and assessment 42% 22% 21% 15%
Student behavior 36% 24% 21% 20%
Strategies to manage big classes 10% 19% 27% 44%
Classroom management and
administration 15% 24% 30% 30%
8 Description and analysis of outcome variables analysis
8.1 Properties and reliability
We used different sources from the selected literature to measure the effectiveness of
induction programmes in different ways. As such, we evaluated (1) teachers’ satisfaction with
job contents and context (Hackman and Oldham, 1974; TALIS, 2013); (2) work related basic
needs satisfaction (Van den Broeck et al., 2010); (3) teacher efficacy (Schwarzer and Hallum,
2008); (4) job stress and burnout (Maslach and Jackson, 1981); and (5) teachers’ motivation
toward specific work tasks (Fernet et al., 2008). Figure 5 summarizes these components.
25
Figure 5: Presentation of the components included in the Survey tool
Note: The numbers between brackets refer to those questions with the highest loadings on the components
measured in a particular question (e.g. Question 14 measures Job Satisfaction), and ‘r’ denotes the reliability
statistic of the scale.
8.1.1 Job Satisfaction
Hackman and Oldham (1974) developed a scale for the measurement of overall satisfaction
and happiness with the job. Overall happiness and satisfaction is measured based on different
questions related to job security, pay and other compensation, peers and co-workers (also
referred to as social satisfaction), supervision, and opportunities for personal growth and
development on the job (also referred to as growth satisfaction).
26
Table 5: Scores on the sub-questions of the scale job satisfaction
Score (max.5)
Q(n°) Description 1 2 3 4 5
5 The degree of respect and fair treatment I receive from the
head teacher 3% 5% 9% 45% 38%
6 The feeling of worthwhile accomplishment I get from
doing my job as a teacher 0% 1% 4% 31% 64%
8 The amount of support and guidance I receive at this
school 2% 7% 5% 49% 36%
10 The amount of independent thought and action I can
exercise in my job as a teacher 2% 7% 7% 53% 31%
11 How secure things look for me in the future in this school 2% 4% 10% 41% 44%
14 The overall quality of support and guidance I receive at
this school 3% 8% 7% 51% 31%
Note: The scores range from 1 (extremely dissatisfied) to 5 (extremely satisfied). The value of 3 indicates
neutral.
Table 3 summarizes the sub-questions with the highest loadings on the scale. The reliability
statistic derived from Cronbach’s alpha is equal to 0.822 based on the six questions reported in
Table 3. Hereby, the scale for job satisfaction meets the generally accepted lower bound criteria
of r≥0.7.
8.1.2 Work Related Needs Satisfaction
The work-related basic needs satisfaction (W-BNS) scale of Van den Broeck et al. (2010)
captures three scales, namely: the need for competence (1), autonomy (2) and relatedness (3)
at work. We present the reliability statistic of each of these scales in Figure 5. The reliability
statistic of the needs for competency scale is poor and equal to r=0.268. For the other two scales
we also estimate moderate reliability coefficients of (r=0.599; needs for autonomy) and
(r=0.588; needs for relatedness at work).
27
Table 6: Scores on the sub-questions of the scales needs for competence, autonomy and
relatedness at work
Scores (max.4)
Q(n°) Description 1 2 3 4
Needs for competence
2 I feel confident in teaching 1% 3% 12% 84%
4 When I'm at work I have serious doubts about whether I can do things
well
17% 18% 20% 45%
5 At work, I feel capable to perform my task 0% 1% 6% 93%
7 I feel comfortable in my task as a teacher 3% 8% 23% 66%
Needs for autonomy
8 I feel forced to do many things on my job I wouldn't choose to do 74% 10% 8% 8%
9 I feel a sense of choice and freedom in my job 5% 10% 23% 63%
10 I feel pressured to do too many things on my job 64% 13% 11% 13%
Needs for relatedness
1 I have a good relationship with the other teachers 1% 2% 9% 89%
3 I feel the relationships (with students, teachers, parents, head teacher...) I
have at work are just superficial
0% 4% 25% 71%
6 I am open to share my failures with my fellow teachers 0% 2% 5% 93%
Note: The scores range from 1 (Not at all true) to 4 (Exactly true).
Table 6 summarizes the sub-questions underlying the three W-BNS scales. With regard to
the need for competence, teachers respond very differently to question 4 “When I'm at work I
have serious doubts about whether I can do things well”. The responses to this question seem
to contradict with the overall fair to good responses on the other questions underlying the needs
for competence. Indeed, Cronbach’s alpha rises to (r=0.520) when leaving out question 4, but
this is still lower than the acceptable level.
We also observe a lot of variance in the responses to the questions on the needs for autonomy.
We improve the estimated Cronbach’s alpha to (r=0.647, instead of =0.559) when leaving out
the positive formulated question on “I feel a sense of choice and freedom in my job”. However,
we did not leave out this question as we need at least three questions in a scale, in order for the
scale to reliably represent the overarching concept.
The questions on the needs for relatedness have been answered in favour of relatedness.
Cronbach’s alpha could be improved a lot, however, to (r=0.668) when leaving out the question
on sharing failures with fellow teachers, but, as discussed above, we do not want to delete this
question.
28
8.1.3 Experienced Job Stress and Burnout
In order to measure experienced job stress and burnout we have used two scales provided by
Maslach and Jackson (1981). These scales are: personal accomplishments and emotional
exhaustion. In particular, emotional exhaustion lies at the heart of experienced burnout
(Maslach, Schaufeli, & Leiter, 2001). If burnout is present, people tend to evaluate themselves
negatively and express feelings of unhappiness with their accomplishments on the job.
Furthermore, burnout can lead to depersonalization (e.g. I feel I consider some students as if
they were impersonal ‘objects’). Table 7 summarizes the results on the questions with highest
loadings on the two scales.
Table 7: Scores on the sub-questions of the scales personal accomplishments and emotional
exhaustion
Scores (max.4)
Q(n°) Description 1 2 3 4 Personal accomplishments
2 I deal very effectively with the problems of my
students 1% 7% 18% 73%
3 I feel I am positively influencing children’s lives
through my work 1% 3% 9% 87%
5 I can easily create a relaxed atmosphere with my
children 0% 2% 11% 87%
7 I can accomplish many worthwhile things in my
job as a teacher 1% 1% 9% 89%
8 In my work, I deal with emotional problems very
calmly 2% 6% 18% 75% Emotional exhaustion
1 I feel emotionally tired from my work 62% 32% 4% 1%
2 I feel exhausted at the end of the workday 8% 55% 23% 13%
3 I feel tired when I get up in the morning and have
to face another day on the job 38% 49% 10% 4%
4 Working with children all day gives me stress 76% 21% 2% 0%
5 At the end of the day I am so tired that I want to
cry 59% 32% 5% 3%
6 I feel frustrated by my job 89% 8% 2% 1%
7 I feel I am working too hard on my job 52% 35% 7% 6%
8 I feel that I can’t take it any longer to continue
teaching 91% 8% 0% 1%
Note: For the scale of personal accomplishments, the scores range from 1 (Not at all true) to 4 (Exactly true).
And for the scale of emotional exhaustion, the scores range from 1 (I never experience this) to 4 (I very always
experience this).
29
We have estimated reliability statistics close to or greater than 0.7 for ‘personal
accomplishments’ (r=0.662) and ‘emotional exhaustion’ (r=0.808).
8.1.4 Self-Assessed Teacher Efficacy
According to Schwarzer and Hallum (2008), self-assessed teacher efficacy is an important
predictor for job stress and burnout. The authors use a scale for the measurement of teacher
self-efficacy, which has been constructed in Schwarzer, Schmitz and Daytner (1999). The scale
consists of 10 questions in total. The estimated reliability statistic for teacher efficacy is equal
to (r=0.685). The answers on the questions underlying self-assessed teacher efficacy can be
found in Table 8.
Table 8: Scores on the sub-questions of the scale self-assessed teacher efficacy
Scores (max.4)
Q(n°) Description 1 2 3 4
1 I am convinced that I am able to teach successfully all relevant subject
content to even the most difficult students 3% 13% 27% 57%
2 I know that I can maintain a positive relationship with parents, even when
tensions arise 2% 5% 17% 77%
3 When I try really hard, I am able to reach even the most difficult students 1% 4% 15% 80%
4 I am convinced that as time goes by, I will continue to become more and
more capable of addressing my students' needs 0% 2% 7% 91%
7 If I try hard enough, I know that I can exert a positive influence on both
the personal and academic development of my students 1% 3% 8% 88%
8 In my work, I am convinced that I can develop creative ways to deal with
system constraints and continue to teach well 2% 5% 18% 74%
Note: The scores range from 1 (Not at all true) to 4 (Exactly true).
8.1.5 Motivation toward Work Tasks
The Work Tasks Motivation Scale for Teachers (WTMST) assesses five motivational
constructs toward work tasks, namely: intrinsic motivation; identified regulation; introjected
regulation; external regulation; and amotivation. Each motivational construct consists of 3
questions so that we have 15 questions in total measuring teachers’ overall motivation toward
work tasks. The reliability statistics are moderate for intrinsic motivation (r=0.718); for
30
introjected regulation (r=0.604); and for external regulation (r=0.607). The other reliability
statistics are poor for identified regulation (r=0.493); and for amotivation (r=0.558). Only
intrinsic motivation meets the critical bound. Table 9 presents the answers to the questions
underlying the different motivation scales.
Table 9: Scores on the sub-questions of the scales intrinsic motivation; identified regulation; introjected
regulation; external regulation; and amotivation
Scores (max.5)
Q(n°) Description 1 2 3 4 5
Intrinsic motivation
1 Because teaching is pleasant to carry
out 2% 1% 5% 19% 74%
6 Because I find teaching interesting to
do 3% 2% 4% 20% 72%
10 Because I like teaching 3% 1% 3% 16% 77% Identified regulation
2 Because teaching enables me to achieve
my own work objectives 1% 3% 5% 25% 66%
12 Because teaching is important for me to
do this task 1% 2% 3% 17% 77%
14 Because I find teaching important for
the academic success of pupils 1% 0% 1% 10% 87% Introjected regulation
4 Because I would feel guilty if would
not teach 14% 4% 8% 20% 54%
7 Because if I don’t teach I will feel bad 23% 7% 12% 23% 35% External regulation
5 Because I’m paid to teach 34% 8% 12% 16% 29%
8 Because my work demands teaching 14% 4% 6% 21% 55%
15 Because the school obliges me to teach 56% 9% 6% 12% 17% Amotivation
3 I don’t know, I don’t always see the
relevance of teaching 82% 5% 6% 4% 3%
11 I used to know why I was teaching, but
I don’t see the reason anymore 78% 5% 8% 5% 5%
13 I don’t know, sometimes I don’t see the
purpose of teaching 89% 3% 4% 2% 2%
Note: The scores range from 1 (Completely disagree) to 5 (Completely agree). The value of 3 indicates neutral.
8.2 Discussion of quality
We have measured several non-cognitive outcome variables by using validated
questionnaires from past published research in the field. Further, we have conducted several
pilot initiatives in order to validate and test the questionnaire. Despite the good overall quality
31
of the selected questionnaires, the extensive piloting phase and the professional
translation/presentation of the English questions into Kinyarwanda, we still observe poor
reliability statistics for some outcome measures. This is most likely due to the population of
study: whereas we conduct the questionnaire among Rwandan teachers, in most cases the
questionnaires have only been validated in the past in Western societies, and, additionally, not
necessarily among teachers. Only questionnaire with regard to experienced job stress and
burnout of Maslach and Jackson (1981), have been explored in African countries (Rajan &
Engelbrecht, 2018). Further, it may be the case that the Rwandan population answered some
questions in a socially desired way. Cultural norms and values can influence the way we
perceive a question, and, as such, how we provide answer to that question. Furnham (1986)
argues that social desirability is a relatively stable, multidimensional trait of persons in very
different situations, and, therefore, that answer patterns even could be used to study a
populations’ view on (ab)normality. We consider this beyond the scope of this study, but this
drawback should be taken into account when analysing further results.
We argue to keep eight outcome measures for further evaluation that have had stable results
in the pilot and baseline study, namely: Job satisfaction; Teacher efficacy; Personal
accomplishments; Emotional exhaustion; and Intrinsic motivation; Need for autonomy; Need
for relatedness; and Amotivation. Five scales (except for need for autonomy; need for
relatedness; and amotivation) have reliability statistics close to, or above the critical value of
0.7. However, we still include the scales of need for autonomy; need for relatedness; and
amotivation since they contribute to the overall picture that we wish to study.
9 Evaluation of the pre-treatment outcomes
Table 10 presents the mean and standard deviation of the outcome variables together with
the mean difference between the two different treatment groups and the control group. T-tests
are used for computation of significance levels. The mean is the average of each scale
32
(unstandardized coefficients). We only present T-test statistics for those outcomes that satisfy
the quality criteria of being a valid and reliable scale (Section 8.2).
Table 10: T-test of outcome variables
Control group Treatment Group 1 Sig. Treatment Group 2 Sig.
Mean Std.Dev Mean Std.Dev Mean Std.Dev
Job satisfaction 4.2 0.7 4.2 0.7 4.1 0.7
Teacher efficacy 3.7 0.4 3.7 0.4 3.6 0.4 *
Personal
accomplishments 3.8 0.3 3.7 0.4 3.7 0.4
Emotional exhaustion 1.6 0.5 1.5 0.4 * 1.5 0.4
Intrinsic motivation 4.6 0.7 4.6 0.7 4.7 0.6
Needs for autonomy 3.4 0.8 3.4 0.7 3.4 0.7
Needs for relatedness 2.2 0.4 2.2 0.4 2.1 0.4
Amotivation 1.4 0.7 1.5 0.8 1.4 0.7
Note: The maximum score on each scale is 5. Significant levels are denoted at 1%-level (***); 5%-level (**);
and 10%-level (*).
All scales except for emotional exhaustion and needs for relatedness present medium to high
scores. There is no significant difference between the treatment groups and the control group
in this regard. However, the standard deviation for job satisfaction, need for autonomy and
intrinsic motivation is quite high (0.7). This indicates that non-negligible variance in answers
to these outcome variables is present. Smaller variance is found for teacher efficacy and
personal accomplishments. The average scores (of 3.7) on these scales are moderately high for
teacher efficacy and personal accomplishments, respectively. We do not observe a significant
difference between the Treatment Group 1 and the control group on these two scales. There is
a significant difference at 10%-level on teacher efficacy between Treatment Group 2 and the
control group. Similarly, we observe a significant difference at 10%-level between Treatment
Group 1 and the control group on emotional exhaustion. Overall, we can conclude that the RCT
design succeeds well in making the treatment groups and control group comparable.
33
10 Conclusion
This paper presented a baseline study on the evaluation of a newly developed mentoring
certificate programme in Rwanda. The objective of this baseline study is to assess the baseline
status of primary outcome measures for the intervention and control group and test for any
differences between the two groups. The primary outcome measures are job satisfaction, work
related needs satisfaction, job stress and burnout, teacher efficacy and motivation. Furthermore,
the baseline results on these outcome measures are described.
In order to conduct the baseline study, questionnaires have been administered among a
sample of 719 teachers. Of these respondents, 412 teachers will eventually receive the newly
established mentoring certificate programme. As such, the baseline study refers to the period
before the execution of this programme.
This baseline study shows that overall treatment and control groups are very similar on pre-
treatment characteristics and on the baseline outcome measures.
11 References
Algozzine, B., Gretes, J., Queen, A. J., & Cowan-Hathcock, M. (2007). Beginning teachers'
perceptions of their induction program experiences. The Clearing House: A Journal of
Educational Strategies, Issues and Ideas, 80(3), 137-143.
Broeck, A., Vansteenkiste, M., Witte, H., Soenens, B., & Lens, W. (2010). Capturing
autonomy, competence, and relatedness at work: Construction and initial validation of
the Work‐related Basic Need Satisfaction scale. Journal of Occupational and
Organizational Psychology, 83(4), 981-1002.
Cabus, S.J., Haelermans, C., Flink, I., Peraer, J. and Gasozintwali, A. (2019). Better
Teachers, Better Students? The Role of Induction Programmes for Beginning Teachers.
Unpublished Manuscript of KU Leuven, Research Institute for Work and Society
(HIVA) & VVOB.
Evertson, C. M., & Smithey, M. W. (2000). Mentoring effects on proteges' classroom
practice: An experimental field study. The Journal of Educational Research, 93(5), 294-
304.
Fernet, C., Senécal, C., Guay, F., Marsh, H., & Dowson, M. (2008). The work tasks
motivation scale for teachers (WTMST). Journal of Career assessment, 16(2), 256-279.
Furnham, A. (1986). Response bias, social desirability and dissimulation. Personality and
individual differences, 7(3), 385-400.
34
Glewwe, P. W., Hanushek, E. A., Humpage, S. D., & Ravina, R. (2011). School resources
and educational outcomes in developing countries: A review of the literature from 1990
to 2010 (No. w17554). National Bureau of Economic Research.
Hackman, J. R., & Oldham, G. R. (1974). The job diagnostic survey: An instrument for the
diagnosis of jobs and the evaluation of job redesign projects (No. TR-4). YALE UNIV
NEW HAVEN CT DEPT OF ADMINISTRATIVE SCIENCES.
Hanushek, E. A. (2011). The economic value of higher teacher quality. Economics of
Education review, 30(3), 466-479.
Hanushek, E. A., & Rivkin, S. G. (2010). Generalizations about using value-added measures
of teacher quality. The American Economic Review, 100(2), 267-271.
Hanushek, E. A., & Woessmann, L. (2017). School resources and student achievement: a
review of cross-country economic research. In Cognitive Abilities and Educational
Outcomes (pp. 149-171). Springer International Publishing.
Ingersoll, R. M., & Strong, M. (2011). The impact of induction and mentoring programs for
beginning teachers: A critical review of the research. Review of educational research,
81(2), 201-233.
Maandag, D. W., Deinum, J. F., Hofman, A. W., & Buitink, J. (2007). Teacher education in
schools: An international comparison. European Journal of Teacher Education, 30(2),
151-173.
Maniraho, J. F., & Christiansen, I. M. (2015). Rwandan grade 6 mathematics teachers’
knowledge. Rwandan Journal of Education, 3(1), 66-76. Maslach, C., & Jackson, S. E. (1981). The measurement of experienced burnout. Journal of
organizational behavior, 2(2), 99-113.
Maslach, C., Schaufeli, W. B., & Leiter, M. P. (2001). Job burnout. Annual review of
psychology, 52(1), 397-422.
Ministry of Education (2016). 2015 Education Statistical Yearbook. Ministry of Education:
Kigali. 123pp.
Ngendahayo, E. & Askell-Williams, H. (2016). Rwanda's new competence-based school
curriculum: New approaches to assessing student learning needed. In D. Curtis & J.
Orrell (Eds), Publishing higher degree research: Making the transition from student to
researcher. Rotterdam: Sense Publishers.
Rajan, S., & Engelbrecht, A. (2018). A cross-sectional survey of burnout amongst doctors in
a cohort of public sector emergency centres in Gauteng, South Africa. African Journal
of Emergency Medicine, 8(3), 95-99. Richter, D., Kunter, M., Lüdtke, O., Klusmann, U., Anders, Y., & Baumert, J. (2013). How
different mentoring approaches affect beginning teachers' development in the first years
of practice. Teaching and Teacher Education, 36, 166-177.
Rockoff, J. E. (2004). The impact of individual teachers on student achievement: Evidence
from panel data. The American Economic Review, 94(2), 247-252.
Rockoff, J. E. (2008). Does mentoring reduce turnover and improve skills of new employees?
Evidence from teachers in New York City (No. w13868). National Bureau of Economic
Research.
35
Roehrig, A. D., Bohn, C. M., Turner, J. E., & Pressley, M. (2008). Mentoring beginning
primary teachers for exemplary teaching practices. Teaching and Teacher Education,
24(3), 684-702.
Schwarzer, R., & Hallum, S. (2008). Perceived teacher self‐efficacy as a predictor of job
stress and burnout: Mediation analyses. Applied psychology, 57(s1), 152-171.
Smith, T. M., & Ingersoll, R. M. (2004). What are the effects of induction and mentoring on
beginning teacher turnover?. American educational research journal, 41(3), 681-714.
Stanulis, R. N., & Floden, R. E. (2009). Intensive mentoring as a way to help beginning
teachers develop balanced instruction. Journal of Teacher Education, 60(2), 112-122.
Strong, M. (2005). Teacher induction, mentoring, and retention: A summary of the research.
The New Educator, 1(3), 181-198.
Wang, J., Odell, S. J., & Schwille, S. A. (2008). Effects of teacher induction on beginning
teachers' teaching: A critical review of the literature. Journal of teacher education,
59(2), 132-152.
36